You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Carl Yeksigian (JIRA)" <ji...@apache.org> on 2015/12/07 23:30:11 UTC

[jira] [Commented] (CASSANDRA-9830) Option to disable bloom filter in highest level of LCS sstables

    [ https://issues.apache.org/jira/browse/CASSANDRA-9830?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15045894#comment-15045894 ] 

Carl Yeksigian commented on CASSANDRA-9830:
-------------------------------------------

Do we have any numbers to show memory usage after this change, as well as compaction performance and read performance? I'm looking over the code now, but if we don't already have performance tests, we should have some run.

> Option to disable bloom filter in highest level of LCS sstables
> ---------------------------------------------------------------
>
>                 Key: CASSANDRA-9830
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-9830
>             Project: Cassandra
>          Issue Type: New Feature
>          Components: Compaction
>            Reporter: Jonathan Ellis
>            Assignee: Paulo Motta
>            Priority: Minor
>              Labels: performance
>             Fix For: 3.2
>
>
> We expect about 90% of data to be in the highest level of LCS in a fully populated series.  (See also CASSANDRA-9829.)
> Thus if the user is primarily asking for data (partitions) that has actually been inserted, the bloom filter on the highest level only helps reject sstables about 10% of the time.
> We should add an option that suppresses bloom filter creation on top-level sstables.  This will dramatically reduce memory usage for LCS and may even improve performance as we no longer check a low-value filter.
> (This is also an idea from RocksDB.)



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)